Decision tree on big categorical dataset

datascience.stackexchange https://datascience.stackexchange.com/questions/74675

  •  11-12-2020
  •  | 
  •  

Question

I would like to use algorithm ID3 in order to find a decision tree of my dataset. I would like to see which of the attributes and values lead to the different value of rating (1<= x <= 5). Do you think it is possible to do it? I got stucked but I don't know if it is because it is not a possible thing to do or it is my fault as a beginner. Can you give me some advise or hints please?

Was it helpful?

Solution

You can see this dataset as a regression problem and not a classification one. The label is ordered, so predicting 2.8 is not a mistake, it is just that it is close to 3 (satisfactory but not there).

I can suggest to start with a regression tree and then round the results to have integers.

Also you can consider this a multiclassification task, but as a starter I would go with regression.

Licensed under: CC-BY-SA with attribution
Not affiliated with datascience.stackexchange
scroll top